Partial Data Replay in a Distributed Memory Buffer System

1. A method comprising: sending, on a first data cycle, a first stripe of a first dataset and a second stripe of the first dataset, to a first data buffer and a second data buffer, respectively; receiving first information indicating that: (i) the first stripe of the first dataset was received, at the first data buffer, with no errors; and (ii) the second stripe of the first dataset was received, at the second data buffer, with errors; in response to receiving the first information, sending, on a second data cycle, a first stripe of a second dataset, and the second stripe of the first dataset, to the first data buffer and the second data buffer, respectively; receiving second information indicating that the second stripe of the first dataset was received, at the second data buffer, with no errors; in response to receiving the second information, sending a store command and a target address corresponding to the first dataset; triggering the first data buffer to send the first stripe of the first dataset to a first memory address, based at least in part, on the target address; and triggering the second data buffer to send the second stripe of the first dataset to a second memory address, based at least in part, on the target address; wherein the first stripe of the first dataset comprises: a first portion of the first dataset, an error detection code, a tag that identifies the first portion of the first dataset; wherein the first portion of the first dataset comprises at least one bit of the first dataset; and wherein the first data buffer and the second data buffer are physically separate from one another.